Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

journal homepage: www.elsevier .com/locate /saa

DNA cleavage, antimicrobial studies and a DFT-based QSAR studyof new antimony(III) complexes as glutathione reductase inhibitor

http://dx.doi.org/10.1016/j.saa.2014.10.0301386-1425/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Tel.: +90 312 2021117; fax: +90 312 2122279.E-mail address: nkaracan@gazi.edu.tr (N. Karacan).

Turgay Tunç a, Yasemin Koç b, Leyla Açık c, Mehmet Sayım Karacan d, Nurcan Karacan d,⇑a Ahi Evran University, Science Faculty, Chemistry Department, Kirsehir, Turkeyb Ankara University, Science Faculty, Biology Department, Ankara, Turkeyc Gazi University, Science Faculty, Biology Department, Ankara, Turkeyd Gazi University, Science Faculty, Chemistry Department, Ankara, Turkey

h i g h l i g h t s

� Fourteen new antimony(III)complexes were synthesized.� Their structures were optimized by

DFT/B3LYP/LANL2DZ method.� DNA cleavage, antimicrobial activities

and glutathione reductase inhibitoractivities were screened.� 2D-QSAR analyses were carried out

for glutathione reductase inhibitoryactivities.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 July 2014Received in revised form 16 September 2014Accepted 13 October 2014Available online 19 October 2014

Keywords:Antimony(III) complexesQSARGlutathione reductase inhibitorsDNA cleavage activity antimicrobial activity

a b s t r a c t

New antimony(III) complexes, [Sb(2-aminopyridine)2Cl3] (1a), [Sb(2-aminopyridine)2Br3] (1b),[Sb(5-methyl-2-aminopyridine)2Cl3] (2a), [Sb(5-methyl-2-aminopyridine)2Br3] (2b), [Sb(2-aminopyrimi-dine)2Cl3] (3a), [Sb(2-aminopyrimidine)2Br3] (3b), [Sb(4,6-dimethoxy-2-aminopyrimidine)2Cl3] (4a),[Sb(4,6-dimethoxy-2-aminopyrimidine)2Br3] (4b), [Sb(2-amino-1,3,5-triazine)2Cl3] (5a), [Sb(2-amino-1,3,5-triazine)2Br3] (5b), [Sb(2-guanidinobenzimidazole) Cl3] (6a), [Sb(2-guanidinobenzimidazole)Br3](6b) [Sb(2- benzyl-2-thiopseudeourea)2Cl3] (7a) and [Sb(2- benzyl-2-thiopseudeourea)2Br3] (7b) weresynthesized. Their structures were characterized by elemental analysis, molecular conductivity, FT-IR,1H NMR, LC–MS techniques. Glutathione reductase inhibitor activity, antimicrobial activity and DNAcleavage studies of the complexes were determined. The geometrical structures of the complexes wereoptimized by DFT/B3LYP method with LANL2DZ as basis set. Calculation results indicated that the equi-librium geometries of all complexes have square pyramidal shape. About 350 molecular descriptors (con-stitutional, topological, geometrical, electrostatic and quantum chemical parameters) of the complexeswere calculated by DFT/B3LYP/LANL2DZ method with CODESSA software. Calculated molecular parame-ters were correlated to glutathione reductase inhibitory activity values (pIC50) of all complexes by BestMulti-Linear Regression (BMLR) method. Obtained two-parameter QSAR equation shows that increasein ‘‘maximum partial charge for a H atom’’ and decrease in HOMO–LUMO gap would be favorable forthe glutathione reductase inhibitory activity.

� 2014 Elsevier B.V. All rights reserved.

T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427 1419

Introduction

Glutathione reductase is a key enzyme in the maintenance ofGSH/GSSG ratio, by catalyzing the conversion of oxidized glutathi-one (GSSG) to the reduced form (GSH). This biochemical processesare essential for detoxification of free radicals and reactive oxygenspecies as well as DNA biosynthesis, protection of cells against var-ious oxidative stresses, intra-cellular signal transduction, and generegulation [1]. Glutathione reductase inhibitors are known to pos-sess anticancer and antimalarial activity and they can enhance theeffects of chloroquine or cytotoxic agents as drug sensitizers [2].From a drug development perspective, inhibition of glutathionereductase enzyme is very important. Wyllie and Fairlamb reportedthat Sb(III) complexes, originally used in the treatment of leish-maniasis, inhibits glutathione reductase activity in macrophagesand proposed it as a drug for acute promyelocytic leukemia [3].To date, various organic-based glutathione reductase inhibitorswere published in many articles [4] but a few metal-complex glu-tathione reductase inhibitors have been reported such as Au(I),Ag(I), Sb(III) and ferrocenic complexes [5–7].

Metal-containing drugs have growing importance in therapeu-tics and diagnostics, particularly in anticancer chemotherapy [8].DNA binding properties of metal complexes also gains importanceowing to their diverse applications as diagnostic agents for medicalapplications and cleavage agents for probing nucleic acid structure[9–11]. On the other hand, antimony complexes have variouspotential usage in the field of pharmacy [12,13] such as anthelmin-thic [14,15], antitrypanosomal [16], anticancer [17–20], antileish-manial [21–23] and antimicrobial [24–28] agents.

For the designing of new compounds with enhanced biologicalactivity, quantitative structure-activity relationship (QSAR) has beenan attractive approach by establishing correlation between chemicalstructure modifications and respective changes of biological activity[29]. As part of our ongoing QSAR studies [30–34], in the presentwork, we correlated the glutathione reductase inhibitory activity ofnew antimony(III) complexes So, we synthesized new fourteenantimony(III) complexes of N-donor ligands (2-aminopyridine,C5H6N2; 5-methyl-2-aminopyridine, C6H8N2; 2-aminopyrimidine,C4H5N3; 4,6-dimethoxy-2-aminopyrimidine, C6H9N3O2; 2-amino-1,3,5-triazine, C3H4N4; 2-guanidinobenzimidazole, C8H9N5; 2-ben-zyl-2-thiopseudeourea, C8H10N2S), [Sb(2-aminopyridine)2Cl3] (1a),[Sb(2-aminopyridine)2Br3] (1b), [Sb(5-methyl-2-aminopyridine)2

Cl3] (2a), [Sb(5-methyl-2-aminopyridine)2Br3] (2b), [Sb(2-aminopy-rimi dine)2Cl3] (3a), [Sb(2-aminopyrimidine)2Br3] (3b), [Sb(4,6-dime-thoxy-2-aminopyrimidine)2Cl3] (4a), [Sb(4,6-dimethoxy-2-aminopyrimidine)2Br3] (4b), [Sb(2-amino-1,3,5-triazine)2Cl3] (5a),[Sb(2-amino-1,3,5-triazine)2Br3] (5b), [Sb(2-guanidinobenzimidaz-ole)Cl3] (6a), [Sb(2-guanidinobenzimidazole)Br3] (6b), [Sb(2-ben-zyl-2-thiopseudeourea)2Cl3] (7a) and [Sb(2-benzyl-2-thiopseudeourea)2Br3] (7b). Their structures were characterized by elemental anal-ysis, molecular conductivity, FT-IR, 1H NMR, LC–MS techniques. Glu-tathione reductase inhibitory activity, DNA cleavage activity andantimicrobial activity of all complexes were evaluated and discussed.

Experimental

Materials and methods

All chemicals used in this study were purchased from Sigma–Aldrich. All reagents were used without further purification. TheNMR spectra were recorded in d6-DMSO in a Bruker Ultrashield300 MHz spectrometer. Elemental analyses were determined on aLECO CHNS–932 auto elemental analysis apparatus. The molecularconductivities of the complexes were measured with WTW Cond330i. Infrared spectra were obtained by using a Mattson 1000FT–IR Spectrometer, from 4000–400 cm�1 in KBr pellet. Liquid

Chromatography Mass spectra were obtained by using a PlatformLC–MS with methanol–acetonitrile mixture as the solvent.

Synthesis of the complexes

The complexes were prepared by the following general method[26]. 25 mL methanol solution of ligand was added to antimony(III)halides dissolved in the same solvent in the mole ratio of 2:1 inhydrochloric acid. The mixture was refluxed for 2 days at 60 �C,after that the mixture was concentrated to 1/3 of its initial volumeand allowed to stand for crystallization at room temperature. Theobtained colorless, yellow and pink crystals were filtered and driedin air.

[Sb(2-aminopyridine)2Cl3] (1a)Anal. Calcd. for C10H12Cl3N4Sb: C, 28.85; H, 2.91; N, 13.46.

Found: C, 28.17; H, 2.75; N, 13.97 IR (KBr, m/cm�1): 3442 (mNH2), 1661 (m CCring), 1621 (m CNring), 994 (m CHring). LC–MS(MeOH): m/z [found (calcd)]: 418.34 (416.35) (M + 2): 1H NMR(300 MHz, DMSO-d6): D 2.50 (2H, NH2), 7.95 (H1, Ar), 7.90 (H3,Ar), 6.98 (H4, Ar), 6.86 (H3, Ar), m.p. 174–175 �C, yield: 58.82%.

[Sb(2-aminopyridine)2Br3] (1b)Anal. Calcd. for C10H12Br3N4Sb: C, 21.85; H, 2.20; N, 10.19.

Found: C, 21.04; H, 2.71; N, 10.86. IR (KBr, m/cm�1): 3413 (mNH2), 1658 (m CCring), 1620 (m CNring), 1024 (m CHring). LC–MS:m/z (calcd) = 549.70 (550.62) (M + 1): 1H NMR (300 MHz, DMSO-d6): D 2.50 (2H, NH2), 7.95 (H1, Ar), 7.90 (H3, Ar), 6.98 (H4, Ar),6.86 (H3, Ar), m.p. 177–178 �C, yield: 70.14%.

[Sb(5-methyl-2-aminopyridine)2Cl3] (2a)Anal. Calcd. for C12H16Cl3N4Sb: C, 32.43; H, 3.63; N, 12.61.

Found: C, 33.02; H, 3.82; N, 12.10. IR (KBr, m/cm�1): 3429 (mNH2), 3040 (m CH), 1668 (m CCring), 1625 (m CNring). LC–MS: m/z(calcd) = 444.40 (444.86) (M+). 1H NMR (300 MHz, DMSO-d6): D2.2 (3H, CH3), 3.50 (2H, NH2), 6.95 (H6, Ar), 7.8 (H5, Ar), 7.90 (H1,Ar), m.p. 174–176 �C, yield: 87.12%.

[Sb(5-methyl-2-aminopyridine)2Br3] (2b)Anal. Calcd. for C12H16Br3N4Sb: C, 24.95; H, 2.79; N, 9.70. Found:

C, 25.43; H, 2.38; N, 9.12. IR (KBr, m/cm�1): 3419 (m NH2), 2930 (mCH), 1668 (m CCring), 1627 (m CNring). LC–MS: m/z (calcd) = 579.75(578.98) (M + 1). 1H NMR (300 MHz, DMSO-d6): D 2.3 (3H, CH3),3.50 (2H, NH2), 7.00 (H6, Ar), 7.80 (H5, Ar), 7.90 (H1, Ar), m.p.175–177 �C, yield: 91.47%.

[Sb(2-aminopyrimidine)2Cl3] (3a)Anal. Calcd. For C8H10Cl3N6Sb: C, 22.97; H, 2.41; N, 20.09.

Found: C, 23.39; H, 2.71; N, 20.74. IR (KBr, m/cm�1): 3346 (mNH2), 1658 (m CCring), 1619 (m CNring), 990 (m CHring). LC–MS: m/z(calcd) = 420.32 (419.35) (M + 1). 1H NMR (300 MHz, DMSO-d6):D 5.40 (2H, NH2), 8.40 (H1, H3, Ar), 6.75 (H2, Ar), m.p. 170–171 �C, yield: 62.50%.

[Sb(2-aminopyrimidine)2Br3] (3b)Anal. Calcd. for C8H10Br3N6Sb: C, 17.42; H, 1.83; N, 15.23.

Found: C, 17.96; H, 1.94; N, 15.74. IR (KBr, m/cm�1): 3320 (mNH2), 3155 (m CH), 1684 (m CCring), 1618 (m CNring). LC–MS: m/z(calcd) = 551.68 (551.75) (M+). 1H NMR (300 MHz, DMSO-d6): D5.40 (2H, NH2), 8.60 (H1, H3, Ar), 7.00 (H2, Ar), m.p. 172–173 �C,yield: %72.69.

[Sb(4-6-dimethoxy-2-aminopyrimidine)2Cl3] (4a)Anal. Calcd. for C14H20Cl3N4O4Sb: C, 26.77; H, 3.37; N, 15.61.

Found: C, 27.11; H, 3.67; N, 15.96. IR (KBr, m/cm�1): 3391 (mNH2), 2940 (m CH), 2909 (m CH), 1684 (m CCring), 1656 (m CNring),

1420 T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427

939 (m CHring). LC–MS: m/z (calcd) = 540.43 (539.15) (M + 1). 1HNMR (300 MHz, DMSO-d6): D 3.50 (2H, NH2), 4.90 (3H, OCH3),5.70 (1H, Ar), m.p. >400 �C, yield: 30.24%.

[Sb(4-6-dimethoxy-2-aminopyrimidine)2Br3] (4b)Anal. Calcd. for C14H20Br3N4O4Sb: C, 21.45; H, 2.70; N, 12.51.

Found: C, 21.03; H, 2.17; N, 12.79. IR (KBr, m/cm�1): 3325 (mNH2), 2928 (m CH), 2885 (m CH), 1674 (m CCring), 1658 (m CNring),941 (m CHring). LC–MS: m/z (calcd) = 673.78 (672.35) (M + 1). 1HNMR (300 MHz, DMSO-d6): D 3.60 (2H, NH2), 5.10 (3H, OCH3),5.80 (1H, Ar), m.p. >400 �C, yield: 30.36%.

[Sb(2-amino-1,3,5-triazine)2Cl3] (5a)Anal. Calcd. for C6H8Cl3N8Sb: C, 17.15; H, 1.92; N, 26.66. Found:

C, 17.69; H, 2.08; N, 26.13. IR (KBr, m/cm�1): 1620 (m CCring), 1562(m CNring), 1006 (m CHring). LC–MS: m/z (calcd) = 421.78 (420.30)(M + 1). 1H NMR (300 MHz, DMSO-d6): D 6.80 (2H, NH2), 8.75(1H, Ar), 8.95 (1H, Ar), m.p. 174–176 �C, yield: 30.36%.

[Sb(2-amino-1,3,5-triazine)Br3] (5b)Anal. Calcd. for C6H8Br3N8Sb: C, 13.02; H, 1.46; N, 20.24. Found:

C, 13.79; H, 1.73; N, 21.02. IR (KBr, m/cm�1): 1622 (m CCring), 1563(m CNring), 1008 (m CHring). LC–MS: m/z (calcd) = 554.16 (553.65)(M + 1). 1H NMR (300 MHz, DMSO-d6): D 6.90 (2H, NH2), 8.80(1H, Ar), 8.90 (1H, Ar), m.p. 174–176 �C, yield: 33.26%.

[Sb(2-guanidinobenzimidazole)Cl3] (6a)Anal. Calcd. for C8H9Cl3N5Sb: C, 23.82; H, 2.25; N, 17.36. Found:

C, 23.37; H, 2.02; N 17.13. IR (KBr, m/cm�1): 3430 (m NH2), 3333 (mNH), 3063 (m CHar), 1689 (m CC), 1629 (m CN). LC–MS: m/z(calcd) = 404.08 (403.31) (M + 1): 1H NMR (300 MHz, DMSO-d6):D 11.90 (1H, NH), 8.20 (1H, NH), 7.40 and 7.20 (4H, Ar), 6.40(3H, NH2 and NH), m.p. 217–218 �C, yield: 53.22%.

[Sb(2-guanidinobenzimidazole)Br3] (6b)Anal. Calcd. for C8H9Br3N5Sb: C, 17.90; H, 1.69; N, 13.05. Found:

C, 18.32; H, 1.87; N 12.68. IR (KBr, m/cm�1): 3439 (m NH2), 3324 (mNH), 3071 (m CHar), 1691 (m CC), 1630 (m CN). LC–MS: m/z(calcd) = 538.66 (537.13) (M + 1): 1H NMR (300 MHz, DMSO-d6):D 12.20 (1H, NH), 7.60 (1H, NH), 7.40 (4H, Ar), 7.10 (3H, NH2 andNH), m.p. 219–220 �C, yield: 53.47%.

[Sb(2- benzyl-2-thiopseudeourea)2Cl3] (7a)Anal. Calcd. for C16H18Cl3N4S2Sb: C, 34.40; H, 3.25; N, 10.03; S,

11.48. Found: C, 34.77; H, 3.18; N, 10.43; S, 11.97. IR (KBr, m/cm�1): 3305 (m NH2), 3194 (m NH), 3085 (m CHar), 1637 (m CCar),1385 (m CH). LC–MS: m/z (calcd) = 560.59 (559.51) (M + 1). 1HNMR (300 MHz, DMSO-d6): D 4.50 (2H, CH2S), 7.40 (5H, Ar), 9.30(2H, NH2) and 9.10 (1H, NH), m.p. >400 �C, yield: 54.27%.

[Sb(2-benzyl-2-thiopseudeourea)2Br3] (7b)Anal. Calcd. for C16H18Br3N4S2Sb: C, 27.77; H, 2.62; N; 8.10; S,

9.27. Found: C, 27.22; H, 2.46; N, 7.72; S, 10.08. IR (KBr, m/cm�1):3310 (m NH2), 3206 (m NH), 3091 (m CHar), 1639 (m CCar), 1385 (mCH). LC–MS: m/z (calcd) = 692.94 (692.99) (M+). 1H NMR(300 MHz, DMSO-d6): D 4.50 (2H, CH2S), 7.40 (5H, Ar), 9.20 (2H,NH2) and 9.00 (H, NH), m.p. >400 �C, yield: 77.84%.

Glutathione reductase activity

The activity of the glutathione reductase to reduce the GSSGwas assayed at 25 �C in phosphate buffer (pH 7.2) by monitoringthe oxidation of NADPH at 340 nm. The assays, in a total volumeof 3 mL, contained 2 � 10�4 M NADPH (0.08 mmol), 1 � 10�4 MGSSG (0.04 mmol) and 50 lL (1.0 units/mL) of glutathione reduc-tase solution. The reaction was started by addition of the enzyme

and the decrease of absorbance was measured at 340 nm at every30 s [35]. The same process was repeated with different concentra-tions of the complexes.

DNA cleavage activity

The interactions of the complexes with supercoiled pBR322plasmid DNA were studied by agarose gel electrophoresis. Thecomplexes were incubated with pBR322 DNA in the dark at 37 �Cfor 24 h and electrophoresed in 1% agarose gel electrophoresis.The gel was electrophoresed for 3 h at 60 V in 1XTAE buffer. Afterelectrophoresis, the gel was stained in ethidium bromide, then,DNA viewed with UV-transilluminator.

BamH1 and HindIII restriction enzyme digestion

The complex-DNA mixtures were first incubated for 24 h andthen restricted with 1 Unit of restriction enzyme BamHI and HindIIIfor 1 h at 37 �C. BamHI and HindIII are known to recognize thesequence G/GATCC and A/AGCTT respectively [36]. pBR322 plas-mid DNA contain a single restriction site for each enzymes whichconvert supercoiled form I DNA and singly nicked circular form IIDNA to linear form III DNA. After one hour incubation, therestricted DNA was electrophoresed in 1% agarose gel for 3 h at60 V in TAE buffer. The gel was stained with ethidium bromideand then viewed with a transilluminator and the image captureby a video-camera as a TIFF file.

Antimicrobial activity

The antibacterial activity of the complexes was performedagainst American Type Culture Collection (ATCC) reference bacte-rial strain Staphylococcus aureus (ATCC 25923), Bacillus cereus(NRLL B-3008), Bacillus subtilis (ATCC 29213), Pseudomonas aeru-ginosa (ATCC 27853), Escherichia coli (ATCC 35218 resistant to betalactam antibiotics), E. coli (ATCC 25922 resistant to the antibioticsother than beta lactam antibiotics), Proteus vulgaris (ATCC 8427)and Enterobacter fecalis (ATCC 292112) and fungi Candida albicans(ATCC 10231) and Candida tropicalis (ATCC 13803) by Agar well dif-fusion technique. The nutrient agar medium and Sabouraud dex-trose agar were poured into Petri dishes after solidification teststrain were inoculated in the media. After that a well was madein the Petri dishes by a sterile borer. The complexes (2500 lM)were introduced into the well. Then, all bacterial strain weregrown in nutrient agar medium and incubated at 37 �C for 24 h.The yeast strains were grown in Sabouraud dextrose agar mediumand incubated at 30 �C for 72 h. Chloramfenicol, Ampicillin (anti-bacterial) and Ketoconazole (antifungal) were used as standardantibacterial agents. All testings were repeated for three times,and the mean values were calculated.

Computational details

Quantum calculation were carried out using the B3LYP/LANL2DZ (method/basis) by Gaussian 03 W program [37]. All thestructures were fully optimized. The absence of imaginary frequen-cies verified that all structures were true minima. The output filesfrom Gaussian were transferred to the program CODESSA to calcu-late approximately 350 molecular descriptors (constitutional,topological, geometrical, electrostatic, quantum-chemical). Inaddition, we have used one indicator parameter for presence (1)and absence (0) of chloro antimony(III) complexes.

For regression analysis, glutathione reductase inhibitor activityvalues were converted to negative logarithm scale (pIC50 =�logIC50) and used as dependent variables. Calculated descriptorvalues were used as independent variables. Best Multiple Linear

Fig. 1. Structures of the complexes.

T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427 1421

Regression method in CODESSA software was used to accomplishthe preselection of the descriptors and to build the linear modelequation [38].

The ‘‘breaking point’’ rule was applied to determine the optimalnumber of descriptors in the model equations. This rule is based onthe significant improvement of R2 (DR2 < 0.02–0.04) with respectto the number of descriptors in the model. Consequently, twodescriptors were used as independent variables in our models.

The squared correlation coefficient (R2), leave-one-out cross-validated squared correlation coefficient (R2cv), the Fisher criteria(F), and standard error (s2) were used as criteria for the stabilityand the robustness of the models. The obtained model was alsovalidated with an internal validation method.

Results

Characterization of the complexes

New antimony(III) complexes (Fig. 1) were synthesized asdescribed in Section ‘Experimental’, and characterized by using

elemental analyses, conductometric measurements, LC–MS, FTIRand 1H NMR spectroscopy techniques. The molar conductivity datain DMSO solution show that all complexes are non-electrolyte.According to obtained data, pyridines, pyrimidines and thiopseu-dourea behaves as monodentate ligands with N donor atom on aro-matic ring, however, guanidinobenzimidazole behaves asbidentate ligand with a N donor atom of benzimidazole groupand a N donor atom of guanidine moiety. Geometries of all com-plexes were optimized using DFT/B3LYP/LANL2DZ method/basisset to find most-stable structures. The ground state optimizedgeometry of all complexes (given in Fig. 2) has square-pyramidalgeometry.

Sb(III) complexes with monodentate ligands have a plane withtwo N donor atoms and two Br atoms in trans arrangement, andone Br atom above the plane (in apical position). However, at theantimony(III) complexes with guanidinobenzimidazole ligand (6aand 6b), Sb(III) atom shares a plane with three bromine atomsand one nitrogen atom of benzimidazole group, and one N donoratom of guanidine moiety above the plane (in apical position).

Fig. 2. The ground state optimized structures of the complexes calculated withB3LYP/LANL2DZ method.

1422 T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427

IR spectra of complex 1–5 are very similar to each other. Inthese complexes NH2 stretching vibrations, found about3446 cm�1 in the free ligand, showed no shifting with complex for-mation, showing that ligands did not coordinate to the Sb(III) atomvia N donor atom of NH2 substituent on the ring. The ring stretch-ing vibrations of ligands generally shifted to higher wavenumberwith complex formation, suggesting that ligands bonded to Sb(III)atom via N donor atoms on aromatic ring. These shifting can beexplained by coupling of stretching vibrations of ring with M-Nstretching vibration. m (NH) vibration of guanidinebenzimidazole(6) is shifted towards higher wave numbers from 3188 cm�1 to3333 cm�1 by complex formation, and ring vibrations of complex6 showed significant shifting to higher wavenumber suggestingthat ligand is bonded via nitrogen atom of guanidine moiety andnitrogen donor atom benzimidazole ring in the complex 6a and6b. However, m (NH2) vibration of it showed no shifting with com-plex formation. m (NH) vibration of 2-benzyl-2-thiopseudourea isshifted towards higher wave numbers from 3022 cm�1 to3194 cm�1 by complex formation, suggesting that ligand is bondedvia nitrogen atom of guanidine moiety in the complex 6a and 6b.

However, m (NH2) vibration and ring vibrations of ligand showedno significant shifting with complex formation.

Proton peaks of the ligands 1–5 were shifted to lower field bycomplex formation. Protons of guanidine moiety of ligand 6 appearas broad singlet at 6.90 ppm and complexation cause a upfield shiftabout 0.5 ppm. However, NH proton of benzimidazole group ofligand 6 appearing as singlet peak at 11.00 ppm shifted to down-field with complex formation. 1H NMR spectra of complex 6a and7a were given in Fig. 3. On the other hand, protons of guanidinemoiety of ligand 7 appear as singlet at 98.50 ppm, however, com-plex 7a show, two well-separated singlets at 9.25 and 9.14 ppm,attributed to the NH2 and NH protons respectively.

Mass spectra of all complexes, m/z values were given at Sec-tion ‘Experimental’, showed molecular ion peak of significantintensity with the characteristic peak pattern of Sb, conformingthat all complex were monomeric structure. For example, LC–MSspectra for complex 6a and 7a (given in Fig. 3) display the molec-ular ion peak at m/z 404.08 and 560.59 confirming the presence ofSbLCl3 and SbL2Cl3 molecular formula, respectively.

Inhibition of glutathione reductase

Comparison of IC50 values of all complexes (given in Fig. 4)highlights the following facts:

(a) Complex 6b and 6a exhibited better glutathione reductaseinhibitory activity with IC50 value in the range of 4.87–5.87 lM.

(b) In general, bromoantimony(III) complexes showed higherglutathione reductase inhibitory activity than chloroantimo-ny(III) complexes.

(c) Glutathione reductase inhibitor activity of ligands decreasesas follows: guanidinobenzimidazole > 2-benzyl-2-thiopseu-deourea > triazine > 2-aminopyrimidine > 2-aminopyri-dine > 5-methylpyridine > 4,6-methoxypyrimidine.

(d) Electron donating substituents on aromatic ring decreasethe glutathione reductase inhibitory activity.

(e) More N donor atom on aromatic ring increases the activity.

QSAR analysis

The multilinear regression analysis using Best Multi LinearRegression (BMLR) method for the 14 complexes in the two-parameter model is given in Table 1. In this table, X and DX areregression coefficients of the QSAR equation and their standarderrors, respectively. The models are given in decreasing relevanceorder according to their statistical significance (ordered by t-testvalue). A graphical presentation of the relationship between theexperimental and the predicted pIC50 values for two-parameterQSAR model is given in Fig. 5. Observed and predicted pIC50 values,their difference and molecular parameters in the model equationwere given in Table 2. In the model equation, first descriptor,‘‘max partial charge for a H atom’’ computed by Zefirov is an elec-trostatic descriptor related to charge distribution of H atom [39]. Itis related with hydrogen bond and interaction between the cationand anion. Increase in ‘‘maximum partial charge for a H’’ would befavorable for the inhibition of the glutathione reductase activity ofthe complexes. However, second descriptor, ‘‘HOMO–LUMO energygap’’ is inversely proportional to the inhibition activity. A lowHOMO–LUMO gap implies low stability, easy polarizability forthe molecule and also higher reactivity in the chemical reaction.Molecular orbital surfaces of HOMO/LUMO of the complexes weregiven in Fig. 6.

Absence of collinearity is confirmed by intercorrelation matrixfor the independent variables used in our model. No significantcorrelation was found between two descriptors (�0.0045).

Fig. 3. 1H NMR (above) and LC–MS spectra (below) of complexes 6a and 7a.

Fig. 4. Experimental IC50 values of the complexes.

Fig. 5. Correlation of observed and estimated pIC50.

T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427 1423

Obtained model equation was further tested with an internal vali-dation method (given in Table 3). For this method, initially, the fullset of the 14 structures was sorted in ascending order according tothe pIC50 value and then divided into three subsets A, B and C: the

Table 1The inhibitor activity and molecular parameters of the complexes.

X X

0 �4.3628 � 10�1 2.7721 � 10�1

1 1.9176 � 101 3.35712 �3.8932 � 10�1 4.6510 � 10�2

n = 14 R2 = 0.9133 F = 57.96

first, fourth, seventh structure, etc., formed subset (A); the second,fifth, eight, etc. formed subset (B); and the third, sixth, ninth struc-ture, etc. formed subset (C). Then, these three sets A–C were pre-pared as the combination of two training subsets (A + B), (A + C)and (B + C). The remaining subsets (C, B and A, respectively)become the corresponding test sets. As seen in Table 3, the

t-test Descriptor

�1.5738 Intercept8.7121 Max partial charge for a H atom�5.3707 HOMO–LUMO energy gaps2 = 0.0038 R2

cv = 0.8546

Table 2DFT-based two parameters QSAR model equation by BMLR method.

Compound no. Inhibitor activity Descriptors

pIC50 (exp) pIC50 (calc) DObs-pred. MPCHa EHOMO ELUMO DEHOMO–LUMO

1a 4.85 4.94 �0.09 0.0483 �6.4595 �1.9884 4.47112a 4.67 4.67 0.00 0.0483 �6.2453 �2.5288 3.71663a 4.91 4.81 0.10 0.0484 �6.2981 �1.8909 3.73624a 4.70 4.72 �0.02 0.0484 �6.1545 �2.4183 3.67495a 4.96 5.00 �0.04 0.0522 �6.7860 �2.4368 4.34926a 5.27 5.24 0.03 0.0522 �6.5379 4.3492 4.58987a 4.97 4.87 0.10 0.0539 �6.2472 �2.8629 4.58981b 5.07 5.04 0.03 0.0539 �5.9727 �1.6575 3.86872b 5.05 5.08 �0.03 0.0531 �7.1945 �2.9095 4.28503b 5.10 5.10 �0.00 0.0531 �6.8829 �3.2414 3.64154b 5.06 5.11 �0.05 0.0619 �5.9836 �2.0452 3.93835b 5.17 5.13 0.04 0.0619 �5.8695 �2.2548 3.61486b 5.31 5.33 �0.02 0.0603 �6.3868 �1.8485 4.53847b 5.19 5.21 0.02 0.0603 �6.2342 �2.3764 3.8578

a MPCH: max partial charge for a H atom.

Fig. 6. Molecular orbital surface for the HOMO and LUMO of complexes.

Table 3Internal validation of the QSAR model.

Training set N R2 R2cv F s2 Test set N R2

A + B 10 0.9873 0.9725 255.17 0.0070 C 4 0.9811A + C 9 0.9765 0.9414 283.14 0.0086 B 5 0.9582B + C 9 0.9924 0.9745 261.61 0.0116 A 5 0.9443

1424 T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427

minimum training quality of R2 0.9765 and a minimum predictingquality of 0.9443 demonstrate that the proposed model has asatisfactory statistical stability and validity.

DNA cleavage activity

Change in the mobility of plasmid DNA on agarose gel is com-monly taken as evidence for direct DNA-metal interactions. In anattempt to gain insight into the anti-leishmanial mechanism ofaction of complexes, DNA cleavage activity was investigated bygel electrophoresis. It is well-known that some drugs bind DNA,

changes DNA conformation and damage the DNA. As a result ofthese changes, the rate of DNA migration in an electric field maychange. Plasmid DNA can exist in three forms: as supercoiled circu-lar (Form I), singly nicked relaxed (Form II), and double nicked lin-ear (Form III). Supercoiled form DNA (Form I, SC) migrates faster onthe gel than nicked circular DNA (Form II, NC). If one strand iscleaved, the supercoils will relax to produce a slower moving opencircular form (Form II), if both strands are cleaved, a linear form(Form III) that migrates in between will be generated. In order todetermine whether the complexes cause conformational changesand/or damage to plasmid DNA helix, we studied the complexes

Fig. 7. Electrophoretograms applied to the incubated mixtures of Plasmid DNA (P) and varying concentrations of the compounds (lane 1: 5000 lM, lane 2: 2500 lM, lane 3:1250 lM, lane 4: 625 lM, line 5: 312.5 lM).

T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427 1425

capacity to induce single or double strand breakage of closelycircular DNA.

Fig. 7 gives the electrophoretograms applying incubated mix-tures of plasmid DNA and varying concentrations of the complexes(5000 lM, 2500 lM, 1250 lM, 625 lM, 312.5 lM) after 24 h. LineP applies untreated plasmid DNA as a control. Lines 1–5 apply toplasmid DNA incubated with decreasing concentrations of thecomplexes. Compared with DNA control (Lane P), the patterns ofLanes 2–5, demonstrate that all the complexes are found to exhibitnuclease activity, because all complexes cause conformationalchanges in form I plasmid DNA. The conversion of supercoiled form(Form I, SC) to nicked form (Form II, NC) becomes more efficientwith increasing concentration of complexes 6b, 5b, 3b and 4b,and the emergence of Form III predicts double strand DNA cleav-age. Remaining complexes show that DNA was converted fromsupercoiled form (Form I) to nicked circular form (Form II) without

further conversion to linear form (Form III), predicting singlestrand DNA cleavage. Based upon their ability to convert super-coiled form (Form I, SC) to nicked circular form (Form II, NC) andthen to linear open circular DNA form (Form III, LC), the tendencyof the complexes varies as 6b � 5b > 3b > 4b.

Restriction enzymes cut DNA at or near specific recognitionnucleotide sequences known as restriction sites [40]. BamHI andHindIII are known to recognize the sequences G/GATCC andA/AGCTT respectively [36]. pBR322 plasmid DNA contain a singlerestriction site for each enzyme which convert supercoiled form IDNA and singly nicked circular form II DNA to linear form III DNA.

Fig. 8 gives electrophoretograms of plasmid/complexesrestricted with BamHI enzymes. Lane P applies to untreated andundigested plasmid DNA. Lane P/B applies untreated but digestedDNA with BamHI enzyme, line 1 4 applied to pBR322 plasmidDNA interacted with complexes 6b, 6a, 5a and 3a followed by

Fig. 8. Electrophoretograms for the BamHI (up) and HindIII (down) digestedmixtures of pBR322 plasmid DNA after their treatment with various concentrationsof compounds. (lane P, untreated pBR322 plasmid DNA, and P/B, P/H; pBR322 DNAlinearized by BamHI and HindIII, respectively.)

Table 4Inhibition zones (in mm) of the compounds and reference antibiotics discs against tested microorganisms by disc diffusion method.

Comp. B. subtilis B. cereus E. fecalis S. aureus K. pneumoniae E. coli 35218 E. coli 25922 P. aeroginosa C. albicans C. tropicalis C. krusei

1a 29.00 ± 0.57 26.00 ± 1.00 – – – – – 18.50 ± 0.70 – – –2a 26.00 ± 1.00 25.00 ± 1.00 – – – – – 13.00 ± 1.00 – – –3a 30.00 ± 0.57 25.50 ± 0.70 – – – – – 18.00 ± 0.40 – – –4a 18.00 ± 1.00 20.00 ± 1.00 – – – – – 14.00 ± 1.00 – – –5a 31.00 ± 0.57 26.50 ± 0.70 – – 19.00 ± 1.00 20.00 ± 0.00 18.00 ± 0.00 18.00 ± 0.40 – – –6a 33.33 ± 1.00 30.00 ± 1.00 – – 25.00 ± 0.00 23.67 ± 0.57 20.00 ± 0.56 19.67 ± 0.57 – – –7a 33.00 ± 1.00 29.67 ± 1.10 – – 20.00 ± 1.00 22.67 ± 0.70 19.10 ± 1.00 18.67 ± 0.57 – – –

1b 26.00 ± 1.00 26.00 ± 1.00 – – – – – 12.00 ± 1.00 – – –2b 25.00 ± 1.00 24.00 ± 1.00 – – – – – 11.00 ± 1.00 – – –3b 26.50 ± 0.70 26.00 ± 1.00 – – – – – 11.00 ± 1.00 – – –4b 17.00 ± 1.00 19.00 ± 1.00 – – – – – 12.00 ± 1.00 – – –5b 27.50 ± 0.70 26.50 ± 1.00 – – 18.00 ± 1.00 18.00 ± 0.00 17.00 ± 0.00 – – –6b 30.00 ± 1.00 28.50 ± 0.70 – – 21.00 ± 1.00 22.50 ± 0.70 19.00 ± 0.56 14.00 ± 1.00 – – –7b 26.00 ± 1.00 27.00 ± 1.00 – – 19.00 ± 1.00 22.00 ± 1.00 18.00 ± 1.00 13.00 ± 1.00 – – –Aa 23.00 ± 1.00 – 27.00 ± 1.00 44.00 ± 1.00 – – 18.00 ± 1.00 60.00 ± 0.50 – – –Ca 21.00 ± 0.50 – 20.00 ± 0.51 24.00 ± 1.00 31.00 ± 0.53 8.00 ± 0.52 25.00 ± 0.56 34.00 ± 0.52 – – –Ka – – – – – – – – 11 ± 1 34 ± 2 18 ± 1

a A: Ampicillin, 10 g (2.86 � 10�8 mole), C: Chloramfenicol, 30 g (9.28 � 10�8 mole), Ketoconazole 0.5 mg/mL (9.40 � 10�7 mole).

1426 T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427

BamHI digestion. In our study, DNA complexes mixtures weredigested with BamHI and HindIII enzymes. When plasmid DNAinteracted with complexes 1a, 3a, 5a and 6a followed by BamHIdigestion, two bands were observed for 3a, 6b, 5a except 6a. Theprevention of restriction enzyme BamHI digestion means thatDNA conformation is changed by the complexes interaction. Whenplasmid DNA interacted with the same complexes followed by Hin-dIII digestion, only one band was observed for all of the complexestested except 6a. Complex 6b binds GG and AA bases of DNA. Addi-tionally, complexes 5a binds to AT bases, however, complexes 6aand 3a bind to GG bases of DNA.

In vitro antimicrobial activity

All complexes were tested for their in vitro antimicrobialactivity against some bacteria and fungi (Table 4). At even higherconcentrations, none of the complexes showed antifungal activityagainst C. albicans C. tropicalis and C. krusei. We were surprisedbecause, dichloroantimony(III) pyrazolinates [41] and anti-mony(III) carboxylates [42] were reported to exhibit antifungalactivity, implying that antifungal activity of antimony(III)complexes depends on type of ligands.

All compounds displayed strong antibacterial activity against, Bcereus, B. subtilis and P. aeruginosa. Compounds 5a, 6a, 7a 5b, 6band 7b showed moderate antibacterial activity against K. pneumo-niae, E. coli 35218 and E. coli 25922. Particularly, the activity ofcompound 6a against B. cereus was comparable or higher than thatof the standard drugs (chloramfenicol and ampicillin). It wasexpected because benzimidazole derivatives and their complexesexhibit various bioactivities including antimicrobial activities[43,44]. In general, it can be concluded that antibacterial activityof trichloroantimony(III) complexes have higher than the tribro-moantimony(III) complexes.

Conclusion

In this paper, we report the glutathione reductase inhibitoryactivity, DNA-cleavage activity and antimicrobial activity of newfourteen antimony(III) complexes. Guanidinobenzimidazol andthiopseudourea Sb(III) complexes showed good biochemical activ-ity. Glutathione reductase inhibitory activity and DNA-cleavageactivity of bromo antimony complexes are significantly morepotent than that of chloro antimony complexes, however, chloroantimony complexes show more antimicrobial activity than bromoantimony complexes. The cleavage of plasmid pBR322 DNA exper-iments show that all our antimony(III) complexes serve as metal-lonucleases. 6b complex behaves as an efficient chemicalnuclease for double-strand cleavage of DNA, and binds both G/Gand A/A bases. Two-parameter QSAR model for glutathione reduc-tase inhibitory activity of fourteen antimony(III) complexes showthat decrease in ‘‘maximum partial charge for a H’’ and HOMO–LUMO gap would be favorable for the glutathione reductase inhib-itory activity.

Acknowledgements

This work has been supported in part by The Scientific andTechnological Research Council of Turkey (TUBITAK) Project no:212T089 and by Russian Foundation for Basic Research. We aregrateful to Prof. Dr. Ningur Noyanalpan for helpful comments.

References

[1] A. Meister, M.E. Anderson, Glutathione, Ann. Rev. Biochem. 52 (1983) 711–760.[2] C. Biot, H. Bauer, R.H. Schirmer, E. Davioud-Charvet, J. Med. Chem. 47 (2004)

5972–5983.[3] S. Wyllie, A.H. Fairlamb, Biochem. Pharm. 71 (2006) 257–267.

T. Tunç et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 136 (2015) 1418–1427 1427

[4] T. Muller, L. Johann, B. Jannack, M. Bruckner, D.A. Lanfranchi, H. Bauer, C.Sanchez, V. Yardley, C. Deregnaucourt, J. Schrevel, M. Lanzer, R.H. Schirmer, E.Davioud-Charvet, J. Am. Chem. Soc. 133 (2011) 11557–11571.

[5] A. Citta, E. Schuh, F. Mohr, A. Folda, M.L. Massimino, A. Bindoli, A. Casini, M.P.Rigobello, Metallomics 5 (2013) 1006–1015.

[6] N. Chavain, E.D. Charvet, X. Trivelli, L. Mbeki, M. Rottmann, R. Brun, C. Bioorg,Med. Chem. 17 (2009) 8048–8059.

[7] M.A.L. Blackie, A. Saravanamuthu, A.H. Fairlamb, K. Chibale, Arkivoc 6 (2008)52–60.

[8] A. Casini, J. Reedijk, Chem. Sci. 3 (2012) 3135–3144.[9] A. Panja, Polyhedron 43 (2012) 22–30.

[10] C.-L. Liu, M. Wang, T.-L. Zhang, H.-Z. Sun, Coord. Chem. Rev. 248 (2004) 147.[11] F. Perveen, R. Qureshi, A. Shah, S. Ahmed, F.L. Ansari, S. Kalsoom, S. Mehboob,

Int. Res. J. Pharm. 1 (2011) 1–8.[12] X. Wang, H. Sun, Comprehensive Inorganic Chemistry II, second ed., 2013.[13] M. Navarro, C. Gabbiani, L. Messori, D. Gambino, Drug Discov. Today 15 (2010)

1070–1078.[14] T. Newlove, L.H. Guimarães, D.J. Morgan, L. Alcântara, M.J. Glesby, E.M.

Carvalho, P.R. Machado, Am. J. Trop. Med. Hyg. 84 (2011) 551–555.[15] Y.Z. Voloshin, O.A. Varzatskii, Y.N. Bubnov, Russ. Chem. B+ 56 (2007) 577–605.[16] G.L. Parrilha, R.P. Dias, W.R. Rocha, I.C. Mendes, D. Benítez, J. Varela, H.

Cerecetto, M. González, C.M.L. Melo, J.K.A.L. Neves, V.R.A. Pereira, H. Beraldo,Polyhedron 31 (2012) 614–621.

[17] I.I. Ozturk, C.N. Banti, M.J. Manos, A.J. Tasiopoulos, N. Kourkoumelis, K.Charalabopoulos, S.K. Hadjikakou, J. Inorg. Biochem. 109 (2012) 57–65.

[18] S.K. Hadjikakou, I.I. Ozturk, M.N. Xanthopoulou, P.C. Zachariadis, S. Zartilas, S.Karkabounas, N. Hadjiliadis, J. Inorg. Biochem. 102 (2008) 1007–1015.

[19] B.M. Day, Martyn P. Coles, Peter B. Hitchcock, Eur. J. Inorg. Chem. 5 (2012)841–846.

[20] I.I. Ozturk, O.S. Urgut, C.N. Banti, N. Kourkoumelis, A.M. Owczarzak, M. Kubicki,S.K. Hadjikakou, Polyhedron 70 (2014) 172–179.

[21] F. Frezard, P.S. Martins, M.C.M. Barbosa, A.M.C. Pimenta, W.A. Ferreira, J.E.Melo, J.B. Mangrum, C. Demicheli, J. Inorg. Biochem. 102 (2008) 656–665.

[22] A.J. Magill, Hunter’s Tropical Medicine and Emerging Infectious Disease, ninthed., Leishmaniasis, 2013. pp. 739–760.

[23] M.I. Khan, S. Gul, I. Hussain, M.A. Khan, M. Ashfaq, I.U. Rahman, F. Ullah, G.F.Durrani, I.B. Baloch, R. Naz, Org. Med. Chem. Lett. 1:2 (2011) 1:2.

[24] H.P.S. Chauhan, A. Bakshi, S. Bhatiya, Spectrochim. Acta Part (A) 81 (2011)417–423.

[25] N.V. Sawant, J.B. Biswal, S.S. Garje, J. Coord. Chem. 64 (2011) 1758–1769.[26] N.M. El-Metwaly, Moamen S. Refat, Spectrochim. Acta Part (A) 81 (2011) 519–

528.[27] U.N. Tripathi, J.S. Solankiz, M.S. Ahmadz, A. Bhardwaj, J. Coord. Chem. 62

(2009) 636–644.[28] K. Mahajan, M. Swami, R.V. Singh, Russ. J. Coord. Chem. 35 (2009) 179–185.[29] C. Hansch, A. Leo, Exploring QSAR: Fundamentals and Applications in

Chemistry and Biology, American Chemical Society, Washington, AcademicPress Inc., 1995. Copyright � 1994.

[30] N. Ozbek, S. Alyar, H. Alyar, E. S�ahin, N. Karacan, Spectrochim. Acta Part (A) 108(2013) 123–132.

[31] H.G. Aslan, N. Karacan, Med. Chem. Res. 22 (2013) 1330–1338.[32] H.G. Aslan, S. Ozcan, N. Karacan, Spectrochim. Acta Part (A) 98 (2013) 329–

336.[33] M.S. Karacan, C. Yakan, M. Yakan, N. Karacan, S.K. Zharmukhamedov, A. Shitov,

D.A. Los, V.V. Klimov, S.I. Allakhverdiev, Biochim. Biophys. Acta 187 (2012)1229–1236.

[34] S. Alyar, N. Ozbek, K. Kuzukıran, N. Karacan, Med. Chem. Res. 20 (2011) 175–183.

[35] M.L. Cunningham, M.J.J.M. Zvelebil, A.H. Fairlamb, Eur. J. Biochem. 221 (1994)285–295.

[36] R.J. Roberts, G.A. Wilson, F.E. Young, Nature 265 (1977) 82–84.[37] M.J. Frisch et al., Gaussian Inc, Pittsburgh PA., Gaussian 03 (Revision B.04),

2003.[38] A. Katritsky, M. Karelson, V.S. Lobanov, R. Dennington, T. Keith, CODESSA

2.7.10, Semichem Inc., Shawnee, KS, 2004.[39] A.R. Katritzky, T. Tamm, Y. Wang, S. Sild, M. Karelson, J. Chem. Inf. Comput. Sci.

39 (1999) 684–691.[40] R.J. Roberts, K. Murray, Crit. Rev. Biochem. 4 (1976) 123–164.[41] U.N. Tripathi, J.S. Solanki, M.S. Ahmad, A. Bhardwaj, J. Coord. Chem. 62 (2009)

636–644.[42] F. Montrichard, F. Le Guen, D.L. Martin, E.D. Charvet, FEBS Lett. 442 (1999) 29–

33.[43] H.Z. Zhang, G.L.V. Damu, G. X Cai, C. H Zhou, Eur. J. Med. Chem. 64 (2013) 329–

344.[44] T. Abdel Ghani, Ahmed.M. Mansour, Spectrochim. Acta Part A Mol. Biomol.

Spectrosc. 86 (2012) 605–613.